Revit 2027

[HVAC-Novice]

Is anyone excited about the new features at all? 

Maybe I'm missing something, but this is a release that makes me even consider skipping this version. Unless the SP bring more tangible features. R2026.4 is so stable, I don't see the reason to risk upgrading. And I'm the person who used to upgrade shortly after release (until I got burned by the schema issue).

It looks like Autodesk looked at the "ideas" forum and decided, "let's just do some AI instead, this is the fad of the week anyway". 

And if a feature is called "Preview", does it mean it doesn't work well yet? 

Curious to hear if anyone tried R2027 and thinks the new features are really that great. 

https://help.autodesk.com/view/RVT/2027/ENU/

I plan to upgrade (if at all) end of the year. But from what I see now, there isn't a single killer feature worth risking moving away from stable R2026. 

[apjones]

I wish we had the luxury of making the choice, but when you are a consulting engineering firm, we have to use whatever version the architect is using.  

This means we have to maintain, and be fluent in, all versions that we can legally run.  We have active projects that ae still in 2021 and have not yet been upgraded (again, we can't upgrade until the architect does....) and we have to maintain our templates and library in the oldest version.   

What an administrative nightmare....

[HVAC-Novice]

---

@apjones wrote:

I wish we had the luxury of making the choice, but when you are a consulting engineering firm, we have to use whatever version the architect is using.  
What an administrative nightmare....

---

Fortunately, we are upstream and I can force downstream designers. And usually the new features or bug fixes warrant annual upgrade. I plan to upgrade in December, but often bugs in the old version force me to upgrade sooner. But R2026.4 just seems perfect so far. I don't see a reason to  upgrade soon. Those new features resolve no single problem for me, and I don't know if Autodesk is capable of AI, if they aren't even capable to let me re-number views on a sheet without me using Dynamo. 

I wish they found a way to get rid of versions and make it a rolling release. BUT I also don't think Autodesk can mange that. One problem with working in different version is, if I create a new family, or schedule or some other great idea for a project in R2026, i also would want to implement that in older projects if they haven't bid out yet. Problem is, if that older (but not bid yet) project is in an older version, i either have to manually re-create my new feature, or not take advantage of it. And I do develop a lot of new stuff. 

[RLY_15]

Having only lightly followed the AI/Agentic discussions on the Dynamo side, I'm not really looking forward to that being embedded in core Autodesk products. There is already pretty vocal pushback on these user forums as-is on having to learn supplemental programming languages or API-level information to deal with specific design requirements or workarounds (why isn't it in base Revit, we're paying for a complete software, not everyone at a company should be technically-oriented, etc. etc. etc.). This is swinging harder in the opposite direction by saying 'look how easy it is to come up with your own solutions!'. Personally, I don't mind access to new solutions being made easier since I'm barely angled towards the technical end... but I don't think the standard expectations of the software are really met yet to address people that just want to go into work, do design work, and go home.

IFC parameter mapping tool sounds great. Want to see more of this. The more that model formats can be interchanged safely between products the better for widescale adoption.

I would *love* to sit down with some structural engineers and find out if the last three years worth of design push into structural calculation methods actually reflects/supports their normal workflows. Each update seems saturated with structural changes, but the majority of structural consultants I come across are still using Revit at a minimalistic level. I assume they're waiting for a critical point where enough needs are met to full pivot.

I'm curious about the updates to the HVAC Zoning rebuild (I never quite liked all the custom UIs to work with setpoints, the more information directly viewable in Properties palette or Type editing, the better), but indifferent to the energy modelling guide. But, considering that this means that HVAC modelling has changed again, I'm getting fatigued with keeping track of the current 'correct' workflow. Also concerned about how model updates that jump to 2027 will break. There's already statements in those patch notes that certain things won't transfer to the new version.

[HVAC-Novice]

---

Having only lightly followed the AI/Agentic discussions on the Dynamo side, I'm not really looking forward to that being embedded in core Autodesk products. There is already pretty vocal pushback on these user forums as-is on having to learn supplemental programming languages or API-level information to deal with specific design requirements or workarounds (why isn't it in base Revit, we're paying for a complete software, not everyone at a company should be technically-oriented, etc. etc. etc.).

I'm curious about the updates to the HVAC Zoning rebuild (I never quite liked all the custom UIs to work with setpoints, the more information directly viewable in Properties palette or Type editing, the better), but indifferent to the energy modelling guide. But, considering that this means that HVAC modelling has changed again, I'm getting fatigued with keeping track of the current 'correct' workflow. Also concerned about how model updates that jump to 2027 will break. There's already statements in those patch notes that certain things won't transfer to the new version.

---

I'm using spaces and zone/Equipment parameters based on what Ripple-HVAC does. That seems to be more robust and independent on what Revit decides each year. the whole energy simulation is pointless since no legal body or LEED will recognize Revit results. 

AI can be good, but I have no trust Autodesk can do it. They probably just subscribed to some ChatGPT or so. I'll judge it when I see it, but based on the total disregard of users' wishes, I'm not optimistic. I hope I'm wrong. But today everything is called AI. Last year ever feature was "green", now it is "AI". The owners probably pressed the  CEO to "do AI like everyone else". 

As for Dynamo etc..... all dependencies seem to change every year. This is horrible for maintainers that only code occasionally. A full time developer can handle it. Remember, we are architects and engineers, not full time coders. 

And simple requests from the ideas forum should be part of the software we pay for. Autodesk sells a car without steering wheel, but they give us some wrenches for us to built some sort of steering mechanism. 

They should stop adding more features like AI, or energy, or carbon or structural, and just implement every year 100+ of the top "ideas" that users requested that will benefit most disciplines or make some modules useful. 

And you probably are correct about structural, it likely isn't correct enough to be used for design. And structural engineers likely rely on proven 3rd party tools. Whenever I work with structural engineers, none uses Revit for structural calculations. 

[RLY_15]

Oh, the Issues Management also looks interesting. Ignoring the branding fatigue that is Bim360->AutodeskDocs->FormaDesignCollaboration, a non-Navisworks location to publish coordination topics seems like a no-brainer good value add.

Rule-base numbering also seems good but 100% there will be tripups when the architect sets their Room rules to remove gaps after element deletion, and the corresponding MEP space + any sequential spaces wind up with numbering mismatches. Also curious what the interaction will be with Room numbering versus embedded doors/windows.

[Simon_Weel]

There are some nice things, but overall, I'm not jumping for joy. I haven't been for years, I might add.

We tend to start using the Revit version in the year it's version number shows. So we started using Revit 2026 this year. As @apjones says - you not always have a choice on using a particular version.

AI is inevitable and it could be useful - if it's smart enough. Calling it Autodesk Assistant makes me shiver. That Assistant you have to use / bypass if you want to create a service request isn't exactly helpful...

And Forma of course. My guess the Big Plan is Revit becoming more and more of a sort of front-end for the cloud.

[arq_42]

The only feature that I find somewhat interesting is the "enhaced stairs risers/threads numering".
And honestly, when I saw it on the roadmap I thought they would include it in the Revit 2026.4 update

 (I actually don't understand why they didn't design the tool this way from the beginning)

[blank...]

Upgraded today, just for the sake of being up to date. The single thing that might be useful are detachable leaders for tags. Nothing else. On the contrary, in MEP they added more clicks to do the same thing.

Should we fix and bring stairs, roofs, topography, railings, beams, wall joins, walls and floors joins, dozens of wrong material takeoffs issues, to 21st century?

No, let's cram a useless LLM into it.

This company is run by people who never worked and have no contact with anyone who works.

[mhiserZFHXS]

---

@blank... wrote:

This company is run by people who never worked and have no contact with anyone who works.

---

If any of us wins a billion dollars in the lottery, we need to promise to create a BIM software developed by the construction industry for the construction industry. Get the tech bros and trust funds out of it. 

[GaryOrrMBI]

The irony there is that Revit was first produced by Architects that were tired of "The big Software Companies" limiting themselves to the traditional CAD based apps... so they designed and built the first "Fully Parametric" design tool with objects that knew what they were and to behave accordingly instead of the typical solids and/or linework on layers to designate what they were supposed to represent.

Then the battle began with ADT and Revit (ok, yeah, there were a couple of others as well but, we're talking Revit and Autodesk right now so...), then the term "BIM" was coined, then Autodesk bought Revit, then... well, here we are, saying that we need the very thing that we had in the late 90's all over again 🙂

-G